This project is designed to assess the clinical and brain metabolic progression of age-related mental decline in the autosomal dominant form of familial Alzheimer disease (FAD). Families with autosomal dominant inheritance patterns for AD are rare, but such pedigrees provide information for linkage analysis of the FAD gene(s). Longitudinal investigation of "at-risk" relatives in these pedigrees is an efficient approach for studying early diagnosis and disease course because 50% of relatives eventually develop the disease. Our group already has recruited 21 FAD Pedigrees. with 410 living relatives potentially available for study. We propose to perform annual clinical and W neuropsychological assessments on relatives at risk for FAD. Positron emission tomography (PET) and magnetic resonance imaging scans will also be performed annually to explore relationships relating brain structure to clinical and L) metabolic decline. Our prior studies demonstrated parietal hypometabolism in patient s with mild dementias who later 5 developed probable sporadic (i.e.. nonfairnilial) AD, with more hypometabolism found in early-onset patients. The proposed project aims to replicate such findings in relatives at risk for FAD. We hypothesize that initial cerebral metabolic patterns (determined with PET) will predict clinical and metabolic decline. We also predict that PET data will improve accuracy of clinical diagnoses earl), in the disease course, thus defining neurobiologically homogeneous 5 FAD pedigrees. Our previous research has demonstrated that the genetic marker, apolipoprotein E type 4 allele (APOE-4) is correlated with the increased risk of Alzheimer disease (AD). While a variety of neuropsychological and functional imaging tests have been demonstrated to predict subsequent cognitive decline, such studies are unlikely to identify very early abnormalities because they: (1) assess brain function during a "resting" state when mental activity is poorly controlled and the specific mental processes showing impairment are not activated; (2) often include subjects without genetic risk for subsequent decline: and. most importantly (3) emphasize measures sensitive only to advanced disease and substantial neuronal loss. We propose to develop methods that detect the earliest signs of brain dysfunction when patients would be most likely to benefit from experimental interventions: before the onset of irreversible and debilitating cognitive impairment. 1a. Develop cognitive behavioral and fMRI procedures that demonstrate activation in brain regions associated with neuronal dysfunction and loss in AD. . 1b. Characterize the brain activation and objective behavioral performance seen in subjects with a clinical diagnosis of probable AD and in age-matched controls 2a. Identify and recruit a cohort of subjects at high risk for the subsequent development of AD, on the basis of genetic markers and pedigree. 2b.Characterize the general neuropsychological performance and baseline pattern of activation (using fMRI) and structure-specific brain volume in this population. 2c. Follow the performance in neuropsychological tests in this cohort in a longitudinal study. 3. Retrospect ively evaluate the fMRI structural and neuropsychological data, comparing the performance of those who did and did not show cognitive declines beyond normal age-associated memory impairment (AAMI).